W3C home > Mailing lists > Public > public-media-capture@w3.org > August 2013

Re: RECAP: Conclusion: Cloning and sharing of MediaStreamTracks - worth it?

From: Adam Bergkvist <adam.bergkvist@ericsson.com>
Date: Thu, 22 Aug 2013 08:16:53 +0200
Message-ID: <5215ACD5.6020403@ericsson.com>
To: Tommy Widenflycht <tommyw@google.com>
CC: Stefan HÃ¥kansson LK <stefan.lk.hakansson@ericsson.com>, Harald Alvestrand <harald@alvestrand.no>, "public-media-capture@w3.org" <public-media-capture@w3.org>
On 2013-08-21 15:04, Tommy Widenflycht wrote:
> Chromium have made some design decisions which assumes that a MST can
> only belong to one MS but that is not the reason for bringing this up.
> It's part of the process to have to rewrite code as the spec evolves and
> from an implementation view I don't see that one is harder to implement
> that the other.
> I just saw a major change in described behavior and could not find any
> compelling reasons why that was made.

The reason why this (same MediaStreamTrack instance in more than one 
MediaStream) works is really just a side-effect of the change we did 
where we wanted to make track cloning explicit. The result was that the 
track instance you called addTrack() with was the actual instance that 
was added to the stream; it wasn't cloned under the hood. Hence, the 
same instance can be added more than once to different MediaStream objects.

So to get rid of the behavior, we need to add something to the API that 
prevents it. There were some suggestions how to do it, but we never 
concluded that the limitation was necessary.

Received on Thursday, 22 August 2013 06:17:38 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:26:19 UTC